DEPARTMENT OF HEALTH AND HUMAN SERVICES
Meeting of: Secretary's Council on Health Promotion and Disease Prevention Objectives for
2010
April 23, 1999, Proceedings
DR. SONDIK: Thank you very much.
At a time like this, I can never resist saying-data, it is the final frontier.
[Laughter.]
And it is one of the critical pieces of Healthy People obviously, but it is obviously not the only critical piece.
Let me start with Volume III. Volume III I think is like all that you see in front of you. looking at achievement. We didn't have a Volume III for Healthy People 2000. Things like it kind of developed over the decade, but the recommendation was that we be very clear on what our data requirements were at the beginning of this process, and Volume III spells it out. I do not know if Dr. DuVal has read every page of Volume III, but I have looked at a great deal of it. I cannot say every single page.
Let me just tell you what is in it, because I think it is extremely important. The first 25 pages is a set of technical-always a bad word to use-but clearly critically important points related to data, and it ranges over everything from the populations involved to definitions to age adjustment in, I think, the right kind of detail for anyone who was concerned with data sources for Healthy People. The bulk of the volume consists of looking at all of the objectives and, for each of the objectives, talking about exactly what the measure is, what the numerator is, what the denominator is, what the data source is, the periodicity that has been outlined, and then a variety of comments.
Now some numbers on these objectives, or numbers, I guess you could say. As Earl pointed out to me, the yellow document had 531 objectives. The count in this one was 612. Dr. Harlan added 28 to that, on paper, not counting what was added by everyone sitting around the table. Plus, many of the objectives are multiple objectives, a, b, c, d, e, and so forth. So we are dealing here with something that is well over 700 objectives.
Of these, about 250 are developmental, meaning that they do not have well-defined data sources at this point. Let's say for the sake of argument, because this is what I do the calculations on, say we have 650 objectives; 250 are developmental, 250 use existing data sources and are probably in pretty good shape. There may be an issue, though, with those, concerning periodicity, and there may be an issue concerning timeliness-in other words, how often we want data and then, whether the data is timely.
But for about 150 of these, a very rough estimate, they call on the use of data sources for which those data sources have not been programmed for these items. For example, it may be, let's say, mammography data from the National Health Interview Survey, which is simply not scheduled to collect mammography data, except at this point, roughly speaking, in 2001. Beyond that, it has not been scheduled. In other words, about 150 of these need to be added to or supplemented to or from the existing set of data sources. This requires the department to be coordinated in doing this, and of course it requires resources, and it is something that we need a process for.
Let me continue with the volume. At the back of this volume, there is a description of the major data sources that are projected to be used for Healthy People, and again, it lists the sponsor and the sample design and so forth, and I think it is in really excellent detail. So, in a sense, this becomes the schematic or the template for data collection for Healthy People.
Now, let me mention some implications of this. First, when I look at Volume III-and it is the same for Volume II, as well, but I just see Volume III in more operational form-Volume III implies to me a departmental commitment of everyone sitting around this table to support the data collection in 2010. Not only that, of course, there is the involvement in intervention activities and research activities that go along with this.
I wish I had a figure to tell you exactly what those resources will be, based on this, but I think it is pretty clear from our discussions that we are going to take another look at these 700-plus objectives and see how we can pare them. In that paring, I think it is extremely important that we look at the data that is being called for and the data sources and put together those who are calling for the data with those who have their fingers on the data sources to be able to come up with a plan that will modify, in effect, Volume III in terms of the data that is being called for and when that data can be collected.
One of the points that has not come up today is the focus on select populations, which is what we are calling-focus on particular populations, individual populations, or special populations, but on select populations. The core of these select populations are outlined by race, whether or not the group is Hispanic, by gender, and by income and education. Then there are about eight other categories that are used for various objectives, including disability, sex orientation, insurance, and a variety of others. The implications of this for the individual data sources I do not believe have really been considered, and it is something that over the next few months we need to do, because clearly it is going to change the scope of each of the data collections and, again, change the resources that are going to be called for.
There are some issues that, again, in these first 25 pages of this volume, that are addressed that each of the data sources and those who are recommending that the data be collected need to consider. First, we have the census, and the census this time is going to be collecting racial data, ethnicity data in a different manner than it has in the past. We can get into what kind of implications that is going to have, but let me just tell you why it is different. It is different in that, instead of an individual saying one race that they are or associate with or whatever the right term is, they in fact will be able to put any number of those. This means we have potentially a very large set of categories.
The census has outlined a set of tabulation guidelines, but we are really all going to have to wait until we see what comes out of the census to get a handle on just how the American public looks at itself over the next decade. What is interesting, I think, is that how it looks at itself is going to change over the next decade. We may see relatively little difference in the 2000 census, but I am willing to bet that, as data collection goes on during this period, our reflections about ourselves are going to change, which is going to pose some interesting issues related to trends.
So in terms of select populations, I am concerned about the resources involved, and I am also concerned about some of the fundamental objectives that we have that may be couched in terms of race and ethnicity as we see it today, but in fact this may change over time. But I think as we progress with 2010, we can deal with it.
Let me mention something about the developmental objectives. As I indicated, about 250, say, out of 650 objectives are developmental. For these, we do not have data sources, which means, again, an attendant set of resources needs to be identified.
I encourage you to look again at the front of this volume, plus chapter 5 in tab V of the first volume that is the IOM report. That has a very useful compendium of issues succinctly stated over just a few pages dealing with data, and it relates to-because we are talking about the creation of new data sources-and they point out that one needs to deal with the general quality of the data issues, of self-reported data versus objective data, validity and reliability, the periodicity and the timeliness of the data, and the point that was brought up earlier today about linkage, linkage of national data to State data to local data, which I was going to come to and that is another significant implication of all of this.
So over the next several years, if we proceed with developmental objectives as we propose to proceed with them, these issues are going to need to be addressed. I also want to echo the point that Debbie Maiese has made concerning the value of the developmental objectives. Many of these-not all, by any means-but many of these over the last decade have borne fruit in terms of new data sources and new insights into prevention and disease control. I think the mere fact that we do not have data sources for these objectives should not be the sole criterion for saying that, no, they don't belong in this. I think we really need to look at the potential value for these.
Even saying that, I think it is very important that the point that was made by several people-Lisa, I think, most recently-related to the science base for these objectives, but for all of the objectives, and this has to be underscored. We have to be hard-nosed, I think, and say that these objectives-that we have evidence that these objectives, if achieved, will make a change in morbidity and mortality, the hardest outcomes from this, the most concrete outcomes.
So for me the implications of this are sort of twofold-the developmental objectives. One is, we need to outline a process for the development of these sources, and the responsibility for this, of course, over the next few years. Plus, we need to have the hard-nosed evaluation of these objectives over the relative near term as to whether or not they should continue to be part of Healthy People 2010. But we also need to apply this science base principle to everything that we see in the volume.
The next point has to do with a point that came up as part of the IOM presentation, and I think it was called linkage in that one, the linkage from national down to State, local, and community data. If all of us who present the data from this or are involved with the data, if there is one issue for which we have the flesh flayed from our body it is this one, because we talk about-these are the national figures-I am looking at Dr. Brooks here-and do not in many cases, cannot in many cases, be related to data at the local level or even the State level for that matter.
There is an issue of where data sources can exist. Are they compatible? Are the definitions the same? Where they do not exist, there is just the fundamental question of how you apply national data to the State and the local level-issues of modeling, issues of the types of surveys that can be done locally and how they relate to those at the national level.
This is something that we really have very few resources for in there. There have been some significant efforts. The CDC has had one effort, but by no means-and that was not necessarily directly related to Healthy People. I think it is something that needs to be high on our priority list in terms of the process for implementing Healthy People, and it is something that is going to need to begin early.
There are some very specific recommendations in the IOM report related to Federal support. I think there may be other alternatives, as well, in particular, perhaps developing a cadre of technical support individuals who can work with people across the country on the development of local data sources using the standards that we use at the national level, the definitions that we use at the national level, plus some research on how we can relate these two levels of data sources.
The next point that I want to make concerns something I guess I have alluded to, but has to do with the schedules for agencies like NCHS, but it also relates to, for example, the SEER program in the National Cancer Institute and data sources across the department, in the scheduling of their activities related to the demands from Healthy People 2010. I see from the NCHS point of view an enormous number of requests for data that we have, some relating to surveys that are fairly well-spoken for in terms of available slots, if you will, up to the middle of the next decade, and some to surveys that, while projected, still do not have resources committed to them, like, for example, one of the really critical areas, the National Survey on Family Growth, which has a number of objectives related to it.
I think we need-I know we need to focus much more clearly on the schedule, even cutting down on the objectives. We need to focus much more clearly on the schedules and the constraints and work to balance the quality of the data that we can achieve with appropriate levels of timeliness and periodicity.
I have mentioned supplements. We also have the President's and Secretary's race initiative, which is focusing on six items in particular, six topic areas in particular. You all know what those are: infant mortality, HIV/AIDS, immunizations, cancer, heart disease, and diabetes. The data requirements for that initiative I view as an integral part of these data requirements, and the increased focus that I think we will have on those areas in order to meet our commitment of eliminating disparities. In a sense, I view those six areas as the real test areas, the core areas for eliminating disparities. What we learn by intensively applying efforts in those areas will be able to be applied across the board.
Well, the data requirements in those six areas at this point are still being developed, but clearly are going to overlap and be even more intensive than the ones that we are calling for in Healthy People 2010. So when I put all of this together, it says to me that we need a process over the next several months that will focus in even more intensively than we have in the past. I think just the presence of this volume and, of course, the main volume, Volume II, testifies to the intensity of the effort that we have had.
But I think, now that we see it all together, now we have to really focus on in particular on the data efforts and emerge from this with not only this set of requirements but a plan that will relate to, certainly, the major data sources in the department and then a plan that also will relate to those data sources that are going to need to be developed over the next few years. I think it is a considerable, really significant challenge. But I must say, I am buoyed today by what I have heard from people sitting around the table today and the interest and really the commitment to this.
MS. GILLIAM: I am wondering if there is a need to narrow the definition of what we are calling the developmental objective, because even in our own mental health chapter, I can see there is a big difference between some sources that really are under our control, really should not be difficult to do-it is a developmental objective because we do not now have the data, but it is pretty clear that we could do those-and all the way to things that would require major new data sources. So I am wondering if part of the problem of the apparent number of developmental objectives is a need to narrow what we are now considering to be a developmental objective. I do not know if others have the same situation in your chapters.
DR. SONDIK: Actually, I had that impression as went through them. I thought that they ranged from those in which it is relatively obvious that one could develop that data source, and presumably usually it was pretty obvious who then would be responsible for that data source, as well, to others in which the objectives seemed to make sense, but the data seemed to be more of a much greater reach.
DR. SATCHER: It sounds like something that is intermediate then-some that you are going to have to develop the data source for; it does not exist today but you are going to be able to develop it in time, so it can become then a valid objective that can be measured between now and the year 2010.
MS. GILLIAM: Right, or even that now a distinction could be made between ones that we know we are there on and the ones that are going to be a reach.
DR. SATCHER: If you know you are there, why is it developmental?
MS. GILLIAM: Because we do not yet have baseline data.
DR. SATCHER: Okay, so you are not there.
DR. SONDIK: It is underdeveloped, perhaps.
MS. GILLIAM: And it will take time to report. Even if we had it going now, it might be a year before we had that.
DR. SONDIK: Or it might be an in-process sort of thing. We could classify it in those terms.
DR. SATCHER: Yes, 250 is a lot of developmental objectives. It is scary.
DR. HARRELL: I think if you really wanted to be serious about this, you would say, is it in the budget? Because I can recall at the end of the process for the year 2000 objectives, Dr. McGinnis, who went from agency head to agency head, sat across the table from them, looked them in the eye, and got them to say that they were going to do this, and it did not happen in a number of cases. So it seems to me the money needs to be where the mouth is in terms of developmental objectives, or there is a good chance they will never get developed.
DR. SATCHER: Yes, that was a question I was going to raise with you, Ed, was, are we in a position to really make sure that we have in the budget, beginning with 2001-I guess we would start with 2000, but certainly with 2001-to request what we need to have in order to assure that this data comes on line and is developed?
DR. SONDIK: If we work hard over the next few months, and this process of paring the objectives works hard over the next few months, then I would say that I think we have a good shot for having an estimate of what it is we are going to need. At this point I am concerned clearly about that, but I am also concerned about the load on the existing data sources, as well, which is significant.
Part of this is not just the sources. Let me just say, part of it is a staffing issue throughout the department, to work on this aspect and people to be, as I said, available to work with State and local communities. There are a variety of ways we can do that, through various granting mechanisms, I would think, a number of ways to do it, but we are extremely tight in every agency, and I have contacts with essentially every agency on this. The staff that works on this is very, very small, and the task of putting this together-they all deserve incredible kudos, particularly the staff at the National Center for Health Statistics and the ODPHP staffs, which really have labored long and hard over this.
DR. SATCHER: Very good.
Yes?
MS. MOORE: I have another question. Are we still on developmental?
DR. SATCHER: Peggy?
MS. GILLIAM: One more comment on developmental measures. I really support what was said earlier, that there needs to be some margin for measures that are really important measures, that even if we got a data source halfway through the process that we would not have gotten if we had not included the measure, that is a worthwhile achievement. So I think there needs to be some margin for discussion there for those kinds of measures.
DR. SATCHER: Okay. Is this on the same issue?
MS. HAYNES: Secretary Shalala mentioned at the beginning of this meeting that there would be funds for prevention initiatives associated with Healthy People. I would assume in that context that budget requests for data collection for evaluation of Healthy People could be part of that.
DR. SATCHER: Oh, definitely it would have to be a part of that.
MS. HAYNES: So it would be good to actually identify what kind of funds you need to collect, these variables that are being collected now, and at least put in a budget request for them.
DR. SATCHER: Yes, that is basically what we were saying, that Ed said we could do if we work hard for the next few months.
MS. MOORE: But this raises another issue and that is the stakeholders issue that was discussed earlier. I mean, could there be resources identified outside of the government, private sector funds that could be made available, especially if the Leading Health Indicators could drive the data collection initiatives?
You mentioned the White House initiative on race. You have sort of seen the candidates, Dr. Sondik, of what the Leading Health Indicators might be. Do any of them raise a red flag in your mind as far as data collection is concerned? Because some of them I think you could find resources for outside government to-if there were a mechanism for looking at what might be available.
DR. SONDIK: No, I was actually very pleased with what I have seen. The issue for me is the select populations and being able to tailor those all the way down to the most detailed geographic area. Exactly what that should be, one has to use judgment clearly, but that is the critical problem I think with those leading up to the data.
MS. MOORE: But you and I have discussed, for example, the way the NHANES survey is designed. It is the only survey that has measured data on height and weight so that BMIs can actually be calculated based on measured data, and it does not even go into nearly half of the States. You know, State-by-State information on something like obesity, I think, is going to be very important, and currently we have no mechanism for collecting such data on a State-by-State basis. I think that is a real problem, myself, from a lot of different perspectives, not only in terms of what we do not know about people's health in each State, but also in terms of the politics of getting funding.
DR. SONDIK: I think you said it very well.
DR. SATCHER: Okay, we are going to move on, because I know some of you have to leave, and I want to try to get through.
Preparing to launch. Mary Jo Deering is going to talk with us about how we are going to launch Healthy People 2010, and then we are going to discuss next steps and see if there are any more public comments, even though I have tried to keep it open throughout the meeting, so that we have gone beyond this table.